TSB overestimation

[docsport77]

Hi all,
just tried to play a bit around with GC to eventually make the switch from WKO.

For that I uploaded a bit of data to play around a bit, 98% bike rides, 2% running, no swimming at the time. Bike and running always with power but 70% of the runs without GPS cause of treadmill use.

TSB is GC is much higher, most times double as high as in WKO or more. Thought it maybe has something to do with the zones setup so I checked every single workout for the time I uploaded and the TSS in GC is mostly lower. ATL and CTL are not so far away from each other, will post some examples, would be great if anyone can point a newbie into the right direction, any help is appreciated

Thank you so much

GC same day, Tri PMC

You can see today in the WKO pic, TSB is -17, ATL 71 CTL 64, here are the GC pics (PMC with Coggan metrics is again almost identical)

[amtriathlon]

CTL, and then TSB, takes a long time to stabilize, importing “a bit of data” is likely not enough for a valid comparison, CTL will be underestimated then TSB overestimated.

[docsport77]

Of course, we are talking about 6 weeks, months would be better. I can easily update this and see if there is something changing. I tried several PMC and one provided a similar result as WKO (with the same data, of course) but I am not sure which one it was, will have to try again.
A TSB 100% or more higher while every workout is showing a lower load is still a bit strange to me as all other metrics are not showing those differences

[amtriathlon]

PMC (Coggan) is based on TSS/BikeStress for Bike activities and should give similar results provided several months of data (6 weeks is not enough) and the same FTP settings. For the remaining PMCs, check the base metric they use, the glossary and FAQs may help.

[docsport77]

@amtriathlon will add more data (y). Rest ist all checked, as written, went to every single workout. We have the same data set for the same time range with the same FTP and GC doubles or triples the TSB as the other data are similar (no matter what PMC o try)

@liversedge Thanks for the points Mark, every file and FTP/Zones are checked. data files coming directly from the unit, no spikes. TSS is nearly identical, only TSB is affected for all PMC´s. Will check the preferences what is the point I have missed so far

[liversedge]

The calculations are trivial and have been reviewed many times in the last 10 years.
If you are seeing differences between a PMC based upon BikeStress in WKO vs GoldeCheetah then either:

• you are not seeding the PMC in the same way (see date ranges)
• the parameters used for ATL and CTL are not the same (see Preferences)
• you have different history for FTP changes (see zones)
• the data files are not the same (make sure you are using the same source data)
• the data files have issues that have been adjusted in WKO (see gaps in recording, spikes)

I would suggest your starting point would be to compare the BikeStress values to TSS in WKO and work from there.

[docsport77]

Last question fpr this, is Skiba PMC including running data (thought not), or maybe it is reading running data if power is included?

I have identical values when I use Skibas PMS and TriScore, just to make sure.

And, can I edit the TSB visual to different colors? Got it that way now as I think most of us, color I like, I like bars...

Would love to see maybe 0 to -15 TSB in green, -16 to -30 in orange and so on

Thanks